Diet of Gobius Vittatus (Gobiidae) in the Northern Adriatic Sea M
Total Page：16
File Type：pdf, Size：1020Kb
Diet of Gobius vittatus (Gobiidae) in the northern Adriatic Sea M. Kovacic To cite this version: M. Kovacic. Diet of Gobius vittatus (Gobiidae) in the northern Adriatic Sea. Vie et Milieu / Life & Environment, Observatoire Océanologique - Laboratoire Arago, 2007, pp.27-33. hal-03234882 HAL Id: hal-03234882 https://hal.sorbonne-universite.fr/hal-03234882 Submitted on 25 May 2021 HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. VIE ET MILIEU - LIFE AND ENVIRONMENT, 2007, 57 (1/2) : 27-33 D I E T OF GOBIUS VITTAT U S (GOBIIDAE) IN THE NORTHERN A D R I ATIC SEA M. KOVA I Prirodoslovni muzej Rijeka, Lorenzov prolaz 1, HR-51000 Rijeka, Croatia [email protected] GOBIUS VITTATUS A B S T R A C T. – The Goby Gobius vittatus is a carnivore and generalist which feeds on a wide DIET variety of prey items, particularly on polychaetes, gastropods, copepods, ostracods and SEASONAL CHANGES ONTOGENETIC SHIFT bivalves. The intensity of feeding was lowest in autumn and the diet spectrum was broadest in FEEDING SELECTIVITY s u m m e r. There were high seasonal differences in biomass and number of almost all major prey items. Smaller specimens fed mostly on meiofauna, and large fish preferred macrofauna. T h e breadth of diet increased with fish size. INTRODUCTION Four microhabitats were recognised as possible source of food at positions were G. vittatus was collected: plankton, scyaphilic The striped goby, Gobius vittatus Vinciguerra, 1883 is phyton, photophilic phyton and psammon. The phyton samples a poorly known Mediterranean gobiid species. T h e were scraped from rocky surfaces (0.01 m2) into plastic bags species was considered rare (Tortonese 1975) or very rare with a solid frame. The psammon samples were collected from (Jardas 1985) due to the lack of data. Morphology and an area of 0.01 m2 to the depth of 2 cm by pulling plastic bags habitat of this species are known from only a few papers with a solid frame through the sand. Samples of plankton were published since species description, based on one or two obtained by draining 3 l containers with air from scuba diving collected specimens (Kova i 2004). Heymer & Zander regulator and filling them up with sea water from the water col- (1978) described habitat, diet and morphology on 21 umn 0-0.3 m above the bottom. Phyton and psammon were specimens from Banyuls-sur-Mer (France). The results fixed in 65% alcohol. Sea water from the containers was filtered on diet were restricted to frequency and abundance analy- through a 100 µm mesh and the samples collected on the net ses of food on 17 specimens collected during two summer were also fixed in 65% alcohol. months. Morphology, habitat preferences and diet from Total length (Lt) of all individuals was measured to the near- these papers, which are the only known data on biology est 0.1 mm (later grouped in 5 mm length classes) and wet mass and ecology of the striped goby, were summarized by weighed (W) to the nearest 0.001 g after blotting dry on Miller (1986). The aim of the present research was to pro- absorbent paper. The specimens were dissected under low vide data on diet of G. vittatus, including diet composi- tion, feeding selectivity, and seasonal and ontogenetic diet shifts, and to compare different methods for diet analysis. MATERIALAND METHODS Seven hundred and four specimens of G. vittatus w e r e obtained on three locations (Stara voda, O tro and Selce) in the Kvarner area of the Northern Adriatic Sea, from April 2001 to March 2002 (Fig. 1). All fish were collected between 8 and 20 m depth, using a hand net and anaesthetic quinaldine during S C U B A dives. Twenty specimens were collected at each loca- tion during one dive every month. In two attempts in January on the location O tro, only four specimens were collected due to bad weather and low temperature. Therefore, the total sample of 704 specimens was collected at three locations in twelve months. All specimens were killed by over- a n a e s t h e t i z a t i o n with quinaldine and fixed in 65% alcohol. Both specimens of G. v i t t a t u s and available food were collected during the same dive Fig. 1. – The Kvarner area, Croatia. Collecting sites: (1) Stara in August and September 2001 at each of the three locations. voda, (2) O tro and (3) Selce. 28 M. KOVA I power binocular microscope for the removal of gut. Guts were quencies (Sokal & Rohlf 1995). Two-ways ANOVA was used to dissected and their entire content sorted to relevant taxonomic assess seasonal and ontogenetic differences in IF. Total length units, which were then counted. Sorted prey items and unidenti- and season were considered to be fixed factors. Homogeneity of fiable residue were weighed wet to the nearest 10 µg (Ohaus variance was tested with Levene’s test, and normality was tested AP250D) after blotting dry on absorbent paper. Assigned wet with Kolmogorov-Smirnov test. IF was log-transformed to satis- masses independent of the animal’s size were used for very fy the A N O VA assumptions. The mean and the confidence lim- small animals (Halacaridae, Ostracoda < 1 mm, Copepoda < its are displayed as the antilogarithm of the mean and the confi- 1 mm), estimated as average mass from weighing of larger sam- dence limits of the log-transformed data (Sokal & Rohlf 1995). ple of specimens. Weight of the entire gut content (WG C) was Tu k e y ’s test was used for post-hoc comparisons after A N O VA . calculated as the sum of weight of all prey items and of weight Data analyses were carried out by the Excel 2002 and the SPSS of unidentifiable residue. The samples of available food were 9.0 program. kept in Rose Bengal (1 g in 1 l 65% alcohol) for 24 h. Plankton samples were then sorted to relevant taxonomic units, which were counted. Very light organisms of available food in samples RESULTS of sands and aufwuchs were extracted by a modified elutriation method (Boisseau 1957). The remaining material was checked Diet for larger and heavier organisms. Separated organisms of psam- mon and phyton were then assorted to relevant taxonomic units, The gut contents of the 704 fish (19.2 Lt 54.0) which were counted. contained 26 taxa (Table I). The other material found in The quantitative importance of different prey in the diet was the gut, consisting mostly of sand and bits of shell, expressed as follows: occurence frequency in percentage ( % F) , rarely of algae and foraminifera, was considered unin- number in percentage ( % N), mass in percentage ( % W) (Hynes tentional intake. The most frequently occurring prey 1950, Berg 1979). The subjective point methods (Hynes 1950, items ( % F > 30%) were Copepoda, Gastropoda, Bival- modified by Joyeux et al. 1991) was also applied as an alterna- via, Polychaeta and Ostracoda (Table I). All other taxa tive method for the determination of amount of the food items in appeared in less than 15% of the analysed guts. T h e terms of matter. Results of percentage mass were compared with most numerous food were Copepoda, followed by Gas- point percentages (%P). Points were given to preys after Joyeux tropoda, Bivalvia and Ostracoda, together constituting et al. (1991), Pampoulie & Bouchereau (1996) and Bouchereau almost 3/4 of all specimens in the prey (Table I). T h e & Guelorget (1999). Points for taxa not present in listed papers results based on percentage mass ( % W) were quite dif- were assigned from listed taxa of similar size. The main food ferent from results of percentage number ( % N) and per- index (IM F), modified with wet mass, was calculated to combine centage frequency ( % F). Gravimetrically, Polychaeta, the three methods used ( % F, % N, % W) (Kova i 2001). Feed- ing intensity was investigated using fullness index (IF) (Hureau Table I. – Diet spectrum of G. vittatus and quantitative contribu- 1970). Seasonal changes and ontogenetic shift in the diet tion of items; % F: occurrence frequency; % N: relative number; %W: relative biomass; % P: points in percentage; IM F: main food breadth were calculated using Levin’s standardised index (Bi) index. (Krebs 1989). The index range from 0 to 1; low values indicate diets dominated by few prey items and high values indicate gen- eralist diets (Krebs 1989). Measure of niche overlap was used to describe overlap between diet and prey offer in surrounding microhabitats (Zander & Hagemann 1987). The simplified M o r i s i t a ’s index (Ci k) was calculated to compare overlap between ingested food and available food of four possible food sources (Krebs 1989). The overlap increases as the Morisita’s index increases from 0 to 1. Overlap is generally considered to be biologically significant when the value exceeds 0.60 (Xie e t a l . 2000). The 95% confidence limits of diet overlap and breadth were estimated using the jackknife method (Krebs 1989). Feeding selectivity (S e l) was evaluated according to Shorygin as mended by Berg (1979). Seasonal changes and ontogenetic shift in the diet composition were analysed by % N and % W and in feeding intensity by the fullness index (IF).